Bone mineral density-fracture risk
A commentary in the British Medical Journal that cites a meta-analysis by Deborah Marshall and colleagues highlights some of the problems with using bone mineral density measurements as the sole determinant for antiresorptive therapy. In the meta-analysis (British Medical Journal, 18 May 1996), the researchers analyzed eleven separate studies, published between 1985 and the end of 1994, which looked at bone mineral density measurements and their ability to predict fractures. The researchers also reviewed case control studies of hip fractures published between 1990 and 1994. After comparing the data, the researchers concluded: “Although bone mineral density measurements can predict fracture risk, they cannot identify individuals who will have a fracture, and a screening programme for osteoporosis cannot be recommended.”
In his BMJ Commentary, Professor Terence J. Wilkin (Plymouth Postgraduate Medical School, UK) questions the rising use of dual energy X-ray absorptiometry machines in the UK when bone density measurements “cannot identify individuals who will have a fracture.” He points out that recommendations for preventive treatment are based on the WHO’s definition of osteopenia that arbitrarily defines it as more than 1 standard deviation below the mean for premenopausal white women. Wilkin asserts that bone turnover, not bone density, should be the focus of antiresorptive treatment (hormone replacement and biphosphonate drugs). He says that the positive effect that these treatments have on preventing fracture cannot be attributed to the small gain in bone density that occurs. Since antiresorptive treatments work equally well in the elderly and the middle-aged, he suggests waiting until fractures are more likely to occur, ages 65-70, instead of starting treatment at menopause. He says that “frequency of impact, which rises exponentially with age, is the main risk factor for fracture.”
In response to Wilkin’s commentary, C. W. McGrother agrees that focusing on elderly people who show a specific risk is a good idea since antiresorptive treatments are expensive and have side effects. McGrother suggests using a fracture risk score based on factors such as bone mineral density, body sway, and muscle strength. Using these factors, T Nguyen and colleagues were able to identify “96% and 93% (sensitivities 88% and 81%) of men and women, respectively, who subsequently developed atraumatic fractures”.